Individualized Drugs' Selection by Evaluation of Drug Properties, Pharmacogenomics and Clinical Parameters: Performance of a Bioinformatic Tool Compared to a Clinically Established Counselling Process

Purpose: Inefficacy and safety concerns are main medications' problems, especially in the case of poly-therapies, when drug-drug interactions may alter the expected drug disposition. Ongoing efforts are aimed to establish drug selection processes aimed to preemptive evaluation of a plethora of factors affecting patient's specific drug response, including pharmacogenomic markers, in order to minimize prescription of improper medications. In previous years, we established at the University Hospital Sant'Andrea of Rome, Italy, a Precision Medicine Service based on a multi-disciplinary experts' team. The team is in charge to produce a drug therapy counselling report, including pharmacogenomic, pharmacokinetic and pharmacodynamic considerations. In this study, we aimed to evaluate the performance of this established "manual" process of therapy selection with a novel bioinformatic tool, the Drug-PIN system.

Patients and methods: A total of 200 patients diagnosed with Major Depressive Disorders or a depressive episode in Bipolar Disorder, with at least three previous failed treatments, who underwent pharmacogenomic profiling and therapy counselling in the Sant'Andrea Hospital from 2017 to 2020. The baseline poly-therapy of these patients was re-evaluated and optimized by Drug-PIN. Results of the Drug-PIN poly-therapy evaluation/optimization were compared with the results of the original poly-therapy evaluation/optimization by therapy counselling. To compare the results between the two processes, the risk associated with each poly-therapy was classified as low, moderate, or high.

Results: The number of baseline poly-therapies classified in low-, moderate- or high-risk did not change significantly between manual system or Drug-PIN system. As the counselling process, also the Drug-PIN system produces a significant decrease in the predicted treatment-associated risk.

Conclusion: Drug-PIN substantially replicates the output of the counselling process, allowing a substantial reduction in the time needed for therapy evaluation. Availability of an effective bioinformatic tool for proper drug selection is expected to exponentially increase the actuation of targeted therapy strategies.